Will AI/ML-defined radios become a reality in the near future? In this paper, we introduce the concept of an AI/ML-defined radio - a radio architecture specifically designed to support AI/ML-based optimization and decision-making in communication functions - and depict its promised benefits and potential challenges. Additionally, we discuss a potential roadmap for the development and adoption of AI/ML-defined radios, and highlight the enablers for addressing their associated challenges. While we offer a general overview of the AI/ML-defined radio concept, our focus throughout the paper remains on Wi-Fi, a wireless technology that may significantly benefit from the integration of AI/ML-defined radios, owing to its inherent decentralized management and operation within unlicensed frequency bands.
@article{arxiv.2405.12675,
title = {Towards an AI/ML-defined Radio for Wi-Fi: Overview, Challenges, and Roadmap},
author = {Boris Bellalta and Katarzyna Kosek-Szott and Szymon Szott and Francesc Wilhelmi},
journal= {arXiv preprint arXiv:2405.12675},
year = {2024}
}